While you’re here… help us stay here.

Are you enjoying open access to policy and research published by a broad range of organisations? Please donate today so that we can continue to provide this service.

Conference paper

An analysis of commuting patterns in Sydney, 2006–2011, using spatial interaction models

Cities and towns Commuting Population forecasting Urban planning Sydney
Attachment Size
apo-nid59763.pdf 554.99 KB

Abstract: Following the seminal work by Wilson (1971), Doubly Constrained Spatial Interaction (DCSI) models have been used to analyse population flows in the form of both commuting (O’Kelly and Lee, 2005, O’Kelly, and Niedzielski, 2008) and also migration. In recent work O’Kelly et al (2012) and O’Kelly (2012) have made conceptual and empirical advances in the application of these models to commuting data. Modelling the pattern of commuting yields both measures of emissivity and accessibility via the (constraint) multipliers and also yields implicit rents and wages by location. In this exploratory paper, the more sophisticated DCSI model developed by O’Kelly (2012) is estimated via a computational search algorithm in Matlab to calibrate parameter estimates using Travel Zone based journey to work data from the 2011 ABS Census for the Sydney Greater Metropolitan Area. We use the implicit rents to examine whether Alonso’s rent gradient can be utilised to identify areas of polycentricity. Also spatial patterns of accessibility are analysed, using Exploratory Spatial Data Analysis. Finally we show that Fotheringham (1983) was correct in highlighting the likelihood of distance decay parameters across different destinations exhibiting global spatial dependence, but there is no evidence of global spatial dependence of origin based distance decay parameters.

Publication Details
Peer Reviewed:
Access Rights Type: